AI Infrastructure Architect

Austin, or Remote (with flexibility for timezone overlap)

We're completely reimagining product analytics by flipping the entire industry approach upside down. While everyone else forces you through the painful dance of data extraction, transformation, expensive tooling, and armies of data engineers just to understand your users, we start where your data already lives: your Postgres product database. From there, we build up and out, connecting database records to event streams to behavioral patterns to research insights to your actual codebase in one unified system that eliminates the traditional pipeline nightmare and puts real product intelligence directly in your hands.

As our first AI Infrastructure Architect, you'll build the unbreakable foundation that makes our AI-native analytics revolution possible at enterprise scale. You'll architect data systems that seamlessly handle everything from a startup's first thousand events to enterprises processing billions of records, designing the infrastructure that keeps customer data secure and compliant while delivering lightning-fast insights. This isn't just backend work, you're engineering the critical systems that let us deliver on our promise to eliminate the analytics nightmare, building infrastructure so robust and intelligent that it makes the impossible feel effortless.

Architect for Scale: Design and build database systems that gracefully scale from gigabytes to petabytes while maintaining sub-second query performance

Secure the Foundation: Implement enterprise-grade security, compliance frameworks (SOC 2, GDPR, HIPAA), and data governance that customers can trust with their most sensitive data

Master the Data Stack: Build sophisticated data infrastructure leveraging Postgres optimization, DuckDB for analytics workloads, Apache Iceberg for data lakes, and S3 for scalable storage

Stream at Speed: Design and implement real-time event streaming pipelines that capture millions of user interactions per second with zero data loss

Own the Backend: Architect APIs, authentication systems, and database layers using Supabase, including RLS policies, performance tuning, indexing strategies, and storage optimization

Build AI-Native Infrastructure: Create MCP (Model Context Protocol) servers and Streamable HTTP endpoints that enable seamless AI-to-data communication

Optimize Relentlessly: Monitor, profile, and optimize every layer of the stack for performance, cost-efficiency, and reliability at scale

Pioneer Standards: Establish infrastructure patterns and best practices as we grow from startup to enterprise platform

Infrastructure Mastery: You've built and scaled backend systems that handle massive data volumes in production environments

Database Wizard: Deep expertise in Postgres optimization, indexing, query planning, and scaling strategies; bonus for DuckDB and SQLite experience

Security-First Mindset: You understand data security, compliance requirements, and how to build systems that meet enterprise standards

Supabase Expert: You know Supabase inside and out: auth, storage, RLS, Edge Functions, real-time subscriptions, and performance optimization

API Architect: You design clean, efficient REST APIs and understand HTTP at a deep level

Data Pipeline Guru: Experience building event streaming systems and analytics pipelines that reliably process high-volume data

Cloud Native: Hands-on experience with Vercel & Cloudflare hosting, Supabase backend, AWS (especially S3), data lake technologies like Apache Iceberg, and modern cloud infrastructure

AI Infrastructure Savvy: You understand what AI applications need from their infrastructure and how to build backends that support LLM-powered features

Full-Stack Capable: Comfortable with Node.js, Type Script, React, and Next.js to collaborate effectively across the stack

Problem-Solving Machine: You debug complex distributed systems issues and architect elegant solutions to gnarly infrastructure challenges

Ship Fast, Scale Smart: You balance rapid iteration with building foundations that won't need complete rewrites at scale

Experience with Databricks and/or…